Generalized advantage estimation combined with trust region optimization enables stable neural network policy learning for complex continuous control from raw kinematics.
Real-time reinforcement learning by sequential actor--critics and experience replay
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2015 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
High-Dimensional Continuous Control Using Generalized Advantage Estimation
Generalized advantage estimation combined with trust region optimization enables stable neural network policy learning for complex continuous control from raw kinematics.